92L Update...

92L has not looked near as good today, and I'm now starting to doubt if it will get named... The NHC has lowered development chances from high to medium over the next 48 hours... In my opinion, if 92L is going to make it to Alex, it's going to have to do it by late Tuesday/early Wednesday... At some point on Wednesday, 92L will start to encounter hostile conditions associated with an upper trough... This upper trough will enhance dry air and wind shear (>30 knots)... The combination of dry air/wind shear (>30 knots) will most likely tear 92L apart! I'm honestly a little surprised that 92L didn't make it to a depression today, even though like I stated yesterday, it is so rare to see any development this early in this part of the Atlantic! While conditions have been mostly favorable, climatology says no! I have wondered if initially 92L being so close to the equator didn't allow for it to spin up as quick, because of the lack of coriolis... Now with 92L near 12N this shouldn't be a problem at all, but again it's running out of time to develop!

A few of the models still want to develop 92L into a tropical storm... The 18z HRWF keeps 92L around 30-40 knots for the next 72 hours... The 18z GFDL has 92L reaching 60 knots by 66 hours, yet tears it up 6 hours later, bringing it down to 37 knots... I guess it thinks the upper level trough holds off a little longer... That's possible, but I personally don't see it! Both the 18z GFS and the 12z European is very weak with 92L... The 12z Canadian is still the most aggressive long term, taking 92L into the Lesser Antilles most likely as a weak tropical storm on Friday... Here is some updated (0z) track/intensity model guidance from CSU...

Please keep in mind that these global models (GFS/Canadian/European) really can't give you a good gauge on strength as a forecaster... While computer models in general have done a MUCH better job with tropical cyclone tracks over the last couple of decades, it still lacks way behind on intensity! The resolution is no where near good enough to pick up on subtle mesoscale features that can make the difference between rapid intensification and no intensification! Even the triply-nested movable mesh model (GFDL) is dependent on GFS model data to "fill in the gaps", or basically account for the overall environment around the tropical cyclone! The GFS has so many heat/heat transfer issues (underestimating the strength of the ridge, not bundling energy, not accounting for latent heat release during hurricanes, etc...), that it becomes the GFDL heat/heat transfer issues as well... Right now this is not so much of an issue, but try to keep this in mind as we get into the heart of the season! I almost guarantee you that the GFS will almost always be to the right of actual track because of these heat/heat transfer issues! Especially with major hurricanes approaching the US... We'll have plenty of chances to watch/test this, trust me on that!

Honestly, it comes down to experience and the repetition of looking at these maps for years & years... You start to get a good feel for things as you gain more and more experience... Also, let's be real here for a second, the models have no clue what's going on right now! Until 92L develops a tight circulation and becomes named, the models are going to have a lot of trouble with initialization no matter what resolution capabilities it has! Bad data in, is bad data out! Especially with something that almost never happens in this region of the Atlantic this early in the season... The models can only do so much! There has to be a human element in forecasting!

I'm much bigger into looking at real time charts and trying to figure it out myself... Not only with the tropics, but with weather forecasting in general... Keep in mind that if you make a forecast off a forecast, then you are truly at the mercy of that model... At this point I have basically zero confidence in the models, so again I'm going to rely off my own analysis on real time data... This doesn't mean I'm not following the models for consistency, but I'm no where near confident enough to lock onto even one model yet! If 92L hasn't developed yet, then I personally only give it about 24 to 36 hours tops to make this happen before the upper trough starts to kill it... No matter what happens, to see the African wave train getting going this early in the season is honestly just scary! We are setting the stage for one heck of an Atlantic hurricane season IMO! That's what I take most out of 92L... 2010 Atlantic hyper season here we come!